Distributed Approach for Joint Symbol and Channel Estimation in Heterogeneous Cellular Networks

Abstract

This work considers the uplink of a multiuser heterogeneous cellular network based on direct-sequence code division multiple access (DS-CDMA). Two cases are considered: i) a centralized case, where a base-station (BS) gathers the information of all users; and ii) a distributed case, where micro-BS nodes perform the signal processing in a distributed fashion. A distributed approach for data estimation and detection is studied in this paper for the uplink of this heterogeneous network which employs spreading codes at the transmitters and consensus-based averaging at the micro-BS nodes. We compare the distributed and centralized approaches considering, in both cases, a receiver structure based on a joint trilinear tensor model for channel gains, spreading codes and transmitted symbols of all users. Numerical results show that the performance of the distributed approach is superior to that of the centralized one.